23 Suppurative keratitis

Background

Definition

Suppurative keratitis (SK) is a severe infection of the cornea characterised by focal polymorphonuclear leucocytic infiltration, typically with surrounding inflammatory cells, oedema and an overlying epithelial defect.1 It is usually caused by bacteria or fungi, and rarely by amoebic or viral infection.

Incidence

The incidence of SK is not accurately known.1 Incidence rates for bacterial keratitis (BK), and microbial keratitis (MK), include ulcerative keratitis as well as SK. In the developing world, MK is responsible for at least 1·5 million new cases of unilateral blindness every year.2 It is a rare cause of blindness in the developed world but has significant associated morbidity.3,4 Worldwide the estimated incidence of BK is approximately 2 per 100 000 population.1 The pattern of SK depends on geographic, environmental and climatic factors.4,5 In the UK, bacteria account for over 90% of SK in cool northern climates6,7 and 60% in hot southern climates.8,9 In tropical climates, fungi are responsible for up to 50% of cases and are usually filamentous.10–15

Aetiology

Suppurative keratitis occurs when the ocular defence mechanisms are disrupted.4 This may be caused by contact lens wear, corneal trauma or surgery, post-herpetic corneal disease, corneal anaesthesia, exposure, keratoconjunctivitis sicca, or ocular surface disease.1,16–18 Contact lens wear is the greatest risk factor for bacterial and amoebic keratitis in developed countries.16,19 Contact lens wearers are exposed to increased numbers of microbes, via bacterial biofilms on contact lens cases,20,21 and have reduced resistance to infection, due to corneal hypoxia and trauma.22,23 In developing countries, corneal trauma, particularly agricultural injury, is the commonest risk factor.24,25 In Africa, fungal keratitis has been associated with HIV infection.26

Prognosis

Suppurative keratitis may result in corneal scarring and opacity leading to severe visual disability.4,17,27 Progression of SK can cause corneal perforation, scleritis, endophthalmitis, panophthalmitis, loss of the eye and even potentially fatal cavernous sinus thrombosis.28,29 This is typically rapid in untreated Psuedomonas keratitis30 and slower with fungal keratitis.31 Prompt diagnosis and treatment can improve the prognosis.16

Question

In patients with bacterial SK what is the best topical antibiotic therapy?

The evidence

Nine papers report eight randomised controlled clinical trials18,27,32–37 that investigated antibiotic therapy in bacterial keratitis. A further trial38 examined whether hyaluronate improves the efficacy of fortified antibiotics by prolonging drug contact time via increased viscosity.39 The details of each trial are summarised in Table 23.1. Four trials18,27,33,34 were of sufficient quality. Two trials32,37 were not double-masked. Four trials32,35,37,38 were not large enough, particularly for subgroup analysis. Patients with BK are a diverse group, therefore subgroup analysis is usually required. Significance levels were not provided by all studies. There were no placebo controlled trials.

The inclusion criterion for the trials was patients with BK. Two studies27,33 randomised patients with clinical BK but only analysed a subset of patients who were culture positive.40 One study38 excluded 10 patients with significant amounts of the same bacteria in both eyes. Four studies32,35,37,38 included moderate ulcers to remove the bias associated with having different sized ulcers. Two studies18,27 only included those with a best-corrected visual acuity of 20/200 or better in the involved eye. Exclusion criteria were microbiological evidence of fungal keratitis, patients with only one eye, pregnant women, patients with diabetes and allergy to study medications. One study also

excluded an unknown number of patients who stopped taking the study medication. They did not state which medication was stopped or the reason for discontinuation.

Results

One small and four large RCTs of sufficient quality found that there was little difference in efficacy between fluoroquinolone monotherapy and combination fortified antibiotic therapy in achieving healing (cure) or resolution of BK. One study reported significantly greater toxicity34 and two studies27,33 found significantly more discomfort with fortified antibiotics. In clinically identified cases of BK positive culture rates were around 50%; staphylococci were most commonly isolated. Fluoroquinolone (ciprofloxacin or ofloxacin) monotherapy can replace fortified antibiotic therapy in the management of moderate bacterial keratitis. Trials should continue to determine the most frequent organisms and their sensitivities.

Comment

The inclusion criteria for some trials was clinically suspected BK, while other trials required microbiological confirmation of BK but this has only moderate sensitivity, such that cases may be excluded. Clinical criteria may allow more cases of BK to be included, however, all non-bacterial cases may not be excluded. Corneal swabs and conjunctival scrapings are less reliable than corneal scrapes. Most of the trials used different study medications or a different concentration of the same medication preventing direct comparison of the results. Further, the single centre studies may only be applicable to patients in a similar setting, as SK varies with geographic location and climate. The main outcome measure was epithelial healing; this may not reflect the rate of bacterial killing. Several studies also included clinical features in their outcome measures. One study identified possible confounders, such as a greater number of cases with Pseudomonas in the fortified antibiotic group than the ciprofloxacin group, but did not adjust for them.

Question

In patients with bacterial SK what is the role of topical steroids when used with topical antibiotic therapy?

The evidence

We found one prospective, randomised, controlled clinical trial of 40 patients in which topical antibiotics alone

(n = 19) were compared to topical antibiotics and steroids (n = 21).41 This RCT was unmasked (patients and investigators) and small. Patients with central or paracentral bacterial corneal ulcers severe enough to warrant hospital admission were included. All patients had a corneal scrape for microscopy and culture. Patients with fungal isolates, perforated ulcers or descemetocoeles, underlying viral keratitis, atopic ulcers, no light perception vision, or aged less than 13 years were excluded. Prior to analysis of the healing rate six out of 21 in the steroid group and eight out of 19 in the non-steroid group were excluded due to a persistent epithelial defect, uncontrolled infection or the requirement for other therapy.

Results

There was no statistical difference in the final VAs for the two groups, though both had a statistically significant improvement in VA (P <0·001 for the steroids group and P <0·01 for the non-steroid group, paired analysis by t test). The mean healing rate was not significantly different, 0·36 mm2/day for the steroid group and 0·30 mm2/day for non-steroid group (Student’s t test, no P values given). There were eight complications in the steroid group and 10 in the non-steroid group (no P values given).

Comment

The single RCT conducted was not of sufficient quality to confirm or exclude important clinical effects of steroid drops in bacterial keratitis. The results may not be applicable to patients at other centres, as the inclusion criterion was BK “severe enough to warrant hospital admission” and this will vary between different centres. The trial was too small to confirm or exclude important clinical effects of steroid drops in bacterial keratitis. Important sources of error in this trial were selection bias, observer bias, measurement bias and confounding in the analysis of healing rates. Limited evidence suggested that topical steroids added to topical antibiotic regimens for BK might not delay healing nor increase complications. Further RCTs should be conducted to investigate topical steroid therapy in BK.

A single RCT found significantly faster healing when hyaluronate was added to topical antibiotics. This trial was too small and not of sufficient quality to allow hyaluronate to be recommended for clinical use in MK. Hyaluronate cannot be routinely recommended for use in MK until further RCTs are conducted.

Table 23.1 Clinical trials of antibiotic therapy in bacterial keratitis

isolates: reported

framycetin 0·5% clinical response:

Hospital,

Psuedomonas.

56%

0·4%

isolates: reported

scrapings in saline* n = 13 saline

Oftamologia,

38 1992

S. aureus

Tobramycin 1·5% 5·9 ± 1·5 days

Parma, Italy

no significant less in the Most frequent 3 patients

83%

v 1·5%Tobramycin

difference ofloxacin isolates: (2 ofloxacin)

v therapyfortified

n = 67

and todue

Staphylococcus group thebetween fortified

7 38%days:

effectsside

aureusS. sp,

<0·001) P(

groups2

therapy

28 86%days:

by 6 patients

=0·41) P(

82%

=0·70) P(

(5 fortified therapy)

=0·49) P(

(Continued)

Table 23.1 (Continued)

Most frequent Ciprofloxacin

Clinical 0·3%Ciprofloxacin clinicalOverall ofResolution Treatment

324

28 incentres

.,al etHyndiuk

isolates: group

(2 bottles) n = 160 efficacy (physician clinical signs failures:

USA,the

33 1996

India

Europe, and

Positive Toxicity:

The ofloxacin Moorfields Eye 122 Clinical Ofloxacin 0·3% Proportion cured Proportion

cultures in Ofloxacin

and saline at 14 days: cured at

study group, Hospital,

49/122 (40%) 10·2%

(placebo) n = 59 ofloxacin 62·1% 7 days:

andLondon

34 1997

v frequentMost

ofloxacin

v

Fortified

Manchester

isolates: fortified

cefuroxime 5·0% fortified therapy 52·5%

Royal Eye

Hospital, UK

Panda Dr R Prasad 30 Microbiological Ofloxacin 0·3% Ulcer resolution: Time to Time to Subjective Most frequent None

healing: symptom relief: improvement isolates: reported

and saline ofloxacin

forCentre

35 1999., alet

Staphylococcus VA:in ofloxacin ofloxacin

(placebo) n = 15 93%

Khokkar Ophthalmic

aureus,

BCVA ≥of

±7·8 days1·54

15·0 3·86±

v cefazolinFortified

Services,

36 2000

20/200 in Coagulase

v

10% and fortified therapy days

New Delhi,

fortified therapy all except negative

v valuesP (no87%

1·5%tobramycin

India

cultures in precipitate

Fortified cefazolin symptoms 15·6 ± 8·6 cases:

Buranapongs, Bangkok,

21/41 (51·2%). noted in

ciprofloxacin

days

graded):

5% and

Thailand

37 2000

(Continued)

Table 23.1 (Continued)

Most frequent 17·6% of

66·7%

v ciprofloxacin 1·4%gentamicin

isolates: ciprofloxacin

v fortified

70·6%

n = 24

group. aeruginosa,P.

therapy fortified therapy

v

S. pneumoniae

S. aureus

Most common Burning and

Venkatesh Aravind Eye 217 Microbiological Ofloxacin 0·3% Time to healing: Proportion Biomicroscopic

isolates: stinging most

Ofloxacin 13.7 healed: findings:

n = 112

Hospital,

.,al etPrajna

IndiaMadurai,

18 2001

was

v ciprofloxacin

discontinued

14·4 days Ciprofloxacin

due to side effects. White crystalline precipitate in 20% of ciprofloxacin group

= 0·32)

P( 77%

.0·80)

=P (

NI, not indicated *the saline contained preservatives.

(Continued)

punctate (22)Fusarium

severe-Non 5;day atresponse

keratitissuppurative chlorhexidine

andInstitute

31 1998

epithelial keratopathy in one chlorhexidine patient, who may

have received

v

2 healedulcers

culture.

topical treatment

natamycin group

v

Excluded patients

more frequently than prescribed

with only one eye or diabetes, children under 1 year, perforated corneas,

:ulcers Severe

1·24–2·63;

polymicrobial

infections, and those unwilling to participate or unable to attend for follow up

* RE = relative efficacy

Question

In patients with fungal SK what is the best topical antifungal therapy?

The evidence

We found three prospective, randomised controlled clinical trials on the treatment of fungal keratitis in developing countries24,31,42 (Table 23.2). One trial enrolled only eight participants and was discontinued after interim analysis revealed an extremely poor clinical response to fluconazole.42 It was not stated whether this trial was masked. Of the two remaining trials, one was masked and in the other the clinical staff could not be masked due to the different appearance of the medications. One trial was too small, particularly for subgroup analysis, and did not adequately control for confounders.24 The two trials used different concentrations of natamycin, preventing a direct comparison of the results. These trials were conducted in tropical settings in the developing world such that their results are not transferable to other clinical settings.

Results

The two large RCTs supported the use of chlorhexidine for the treatment of filamentous fungal keratitis when specific antifungal therapy is not available.24,31 One small RCT suggested that topical and oral fluconazole might not be the agent of choice for treatment of fungal keratitis in India, possibly due to the predominance of filamentous fungi.42

Comment

Available RCTs are not of sufficient quality to determine the most appropriate antifungal therapy for use in developing countries where the availability of antifungals is limited. RCTs are needed to establish the most appropriate antifungal therapy in a variety of clinical settings. Until the results of such trials are available therapy should be based on microbiological findings25 and local epidemiological data of fungal type and sensitivity. Limited evidence suggests that chlorhexidine may be used instead of natamycin in tropical settings where the availability of antifungals is limited.

Implications for practice

Patients with SK are a diverse group with different risk factors and infecting organisms. Antibiotic monotherapy can replace combination therapy in bacterial SK.

Implications for research

Further RCTs are needed to address the role of steroids in BK and to determine the best antifungal therapy in a variety of settings. The quality of future clinical studies would be improved by a faster and more reliable method for diagnosis of SK, such as polymerase chain reaction. Larger studies, particularly multi-centre trials, would allow valid subgroup analysis and ensure that study results are more widely applicable. At the same time, trials should continue to gather local data for monitoring and the selection of the appropriate therapy.

References

1.Wilhelmus KR. Bacterial keratitis. In: Pepose JS, Holland GN, Wilhelmus KR, eds. Ocular Infection and Immunity. St Louis: Mosby, 1996, pp. 973–81.

2.Whitcher JP, Srinivasan M, Upadhyay MP. Prevention of corneal ulceration in the developing world. Int Ophthalmol Clin 2002;42:71–7.

3.Dart JKG, Seal DV. The pathogenesis and therapy of Pseudomonas aeruginosa keratitis. Eye 1988;2(Suppl):S46–55.

4.Schaefer F, Bruttin O, Zografos L, Guex-Crosier Y. Bacterial keratitis: a prospective clinical and microbiological study. Br J Ophthalmol 2001;85:842–7.

5.Houang E, Lam D, Fan D, Seal D. Microbial keratitis in Hong Kong: relationship to climate, environment and contact-lens disinfection.

Trans Royal Soc Trop Med Hygiene 2001;95:361–7.

6.Asbell P, Stenson S. Ulcerative keratitis. Survey of 30 years’ laboratory experience. Arch Ophthalmol 1982;100:77–80.

7.Coster DJ. Inflammatory disease of the outer eye. Trans Ophthalmol Soc UK 1979;99:463–80.

8.Liesegang TJ, Forster RK. Spectrum of microbial keratitis in South Florida. Am J Ophthalmol 1980;90:38–47.

9.Katz NNK, Wadud SA, Ayazuddin M. Corneal ulcer disease in Bangladesh. Ann Ophthalmol 1983;15:834–6.

10.Rahman MM. Management of fungal corneal ulcer. Trans Ophthalmol Soc Bangladesh 1981;9:12–19.

11.Williams G, Billson F, Husain R, Howlader SA, Islam N, McClellan K. Microbiological diagnosis of suppurative keratitis in Bangladesh. Br J Ophthalmol 1987;71:315–21.

12.Dunlop AAS, Wright ED, Howlander SA et al. Suppurative corneal infection in Bangladesh: a study of 142 cases examining the microbiological diagnosis, clinical and epidemiology features of bacterial and fungal keratitis. Aust NZ J Ophthalmol 1994;22:105–10.

13.Willams G, McClellan K, Bilson F. Suppurative keratitis in rural Bangladesh; the value of gram stain in planning management. Int Ophthalmol 1991;15:131–5.

14.Hagan M, Wright E, Newman M, Dolin P, Johnson G. Causes of suppurative keratitis in Ghana. Br J Ophthalmol 1995;79:1024–8.

15.Yee RW, Kosrirukvongs P, Meenakshi S, Tabbara KF. Fungal keratitis. In: Tabbara KF, Hyndiuk RA, eds. Infections of the Eye, 2nd edn. Little, Brown and Company, 1995.

16.Dart JKG, Stapleton F, Minassian D. Contact lenses and other risk factors in microbial keratitis. Lancet 1991;338:650–3.

17.Diamond JP, White L, Leeming JP, Bing Hoh H, Easty DL. Topical 0·3% ciprofloxacin, norfloxacin and ofloxacin in treatment of bacterial keratitis: a new method for comparative evaluation of ocular drug penetration. Br J Ophthalmol 1995;79:606–609.

18.Venkatesh Prajna N, George C, Selvaraj S, Lu KL, McDonnell PJ,

Srinivasan M. Bacteriologic and clinical efficacy of ofloxacin 0·3% versus ciprofloxacin 0·3% ophthalmic solutions in the treatment of patients with culture-positive bacterial keratitis. Cornea 2001;20: 175–8.

19.Duguid IGM, Dart JK, Morlet N et al. Outcome of Acanthamoeba keratitis treated with polyhexamethyl biguanide and propamidine. Ophthalmology 1997;104:1587–92.

20.Elder MJ, Stapleton F, Evans E, Dart JKG. Biofilm-related infections in ophthalmology. Eye 1995;9:102–9.

21.Stapleton F, Dart J. Psuedomonas keratitis associated with biofilm formation on a disposable soft contact lens. Br J Ophthalmol 1995;79:864–5.

22.Dart JKG. Contact lens and prosthesis infections. In: Jaeger E, Tasman W, eds. Duane’s Foundations of Clinical Ophthalmology Vol 2. Philadelphia: Lippincott-Raven, 1996.

23.Liesegang TJ. Contact lens-related microbial keratitis: Part I: Epidemiology. Cornea 1997;16:125–31.

24.Rahman MR, Minassian D, Srinivasan M, Martin MJ, Johnson GJ. Trial of chlorhexidine gluconate for fungal corneal ulcers.

Ophthalmic Epidemiol 1997;4:141–9.

25.O’Day DM. Fungal keratitis. In: Pepose JS, Holland GN, Wilhelmus KR, eds. Ocular Infection and Immunity. St Louis: Mosby, 1996, p. 1048.

26.Mselle J. Fungal keratitis as an indicator of HIV infection in Africa. Trop Doct 1999;29:133–5.

27.O’Brien TP, Maguire MG, Fink NE, Alfonso E, McDonnell P and the Bacterial Keratitis Study Research Group. Efficacy of ofloxacin v cefazolin and tobramycin in the therapy for bacterial keratitis. Arch Ophthalmol 1995;113:1257–65.

28.Brenner M. Bacterial corneal ulcer, endophthalmitis, and embolic phenomena. Ann Ophthalmol 1984;16:334–40.

29.Liesegang TJ. Bacterial keratitis. Infect Dis Clin North Am 1992;6:815–29.

30.Van Bijsterveld OP, Jager GV. Infectious diseases of the conjunctiva and cornea. Curr Opin Ophthalmol 1996;7:65–70.

31.Rahman MR, Johnson GJ, Husain R, Howlader SA, Minassian DC. Randomised trial of 0·2% chlorhexidine gluconate and 2·5% natamycin for fungal keratitis in Bangladesh. Br J Ophthalmol 1998;82:919–25.

32.Reddy PR. Topical antibiotics in the management of corneal ulcer. Ind J Ophthalmol 1988;36:95–7.

33.Hyndiuk RA, Eiferman RA, Caldwell DR et al., the Ciprofloxacin

Bacterial Keratitis Study Group. Comparison of Ciprofloxacin ophthalmic solution 0·3% to fortified tobramycin-cefazolin in treating bacterial corneal ulcers. Ophthalmology 1996;103:1854–63.

34.The Ofloxacin Study Group. Ofloxacin monotherapy for the primary treatment of microbial keratitis. A double-masked, randomised, controlled trial with conventional dual therapy. Ophthalmology 1997;104:1902–9.

35.Panda A, Ahuja R, Srinivas S. Comparison of topical 0·3% ofloxacin with fortified tobramycin plus cefazolin in the treatment of bacterial keratitis. Eye 1999;13:744–7.

36.Khokhar S, Sindhu N, Mirdha BR. Comparison of topical 0·3% ofloxacin to fortified tobramycin-cefazolin in the therapy of bacterial keratitis. Infection 2000;28:149–52.

37.Kosrirukvongs P, Buranapongs W. Topical ciprofloxacin for bacterial corneal ulcer. J Med Assoc Thail 2000;83:776–82.

38.Gandolfi SA, Massari A, Orsoni JG. Low-molecular-weight sodium hyaluronate in the treatment of bacterial corneal ulcers. Graefe’s Arch Clin Exp Ophthalmol 1992;230:20–3.

39.Adler CA, Maurice DM, Paterson ME. The effect of viscosity of the vehicle on the penetration of fluorescein into the human eye. Exp Eye Res 1971;11:34–42.

40.Baker RS, Flowers CW Jr, Casey R, Fong DS, Wilson MR. Efficacy of ofloxacin versus cefazolin and tobramycin in the therapy for bacterial keratitis. Arch Ophthalmol 1996;114:632–3.

41.Carmichael TR, Gelfand Y, Welsh NH. Topical steroids in the treatment of central and paracentral ulcers. Br J Ophthalmol 1990;74:528–31.

42.Rao SK, Madhavan HN, Rao G. Fluconaxole in filamentous fungal keratitis. Cornea 1997;16:700.

Even though the incidence of uveitis is much lower than many other blinding conditions, it still represents an important cause of visual loss in individuals of working age and, in some parts of the world, is responsible for blindness on a very large scale. Some causes of uveitis can be clearly identified, especially the infectious ones in which case specific therapy can be instituted and cure can usually be achieved. This is not always as easy as it may seem, as can be seen in the chapter on Toxoplasmosis and Onchocerciasis.

The majority of cases have no specific diagnosis and are considered idiopathic, being driven by an immune-mediated mechanism. These are treated with anti-inflammatory and immunosuppressive agents in an attempt to control the progressive damage caused by the uncontrolled inflammation. Not much is known about many of these conditions and very little evidence exists to support the use of most of the agents

currently used. The chapter entitled Idiopathic sight-threatening uveitis demonstrates this very clearly.

Anterior uveitis represents the most common form of intraocular inflammation seen in practice, but it is interesting to note that very few studies have dealt with this subject properly. On the other hand CMV retinitis, a much more recent problem, has had a large number of very good RCTs to demonstrate the efficacy of antiviral therapy in controlling progression of this blinding condition in HIVinfected patients. This certainly reflects the difference between the two conditons in terms of their impact as a cause of visual loss, but may also reflect the huge financial drive behind the RCTs in CMV retinitis.

Overall, the chapter dealing with the conditions mentioned above highlight the need for properly designed trials, which is the only way to answer the many questions we are still asking in our daily practice.